/home/angelos/stock-analysis/venv_39/lib/python3.9/site-packages/gluonts/json.py:45: UserWarning: Using `json`-module for json-handling. Consider installing one of `orjson`, `ujson` to speed up serialization and deserialization. warnings.warn(
BTC 2020-08-20 00:00:00 2021-12-01 00:00:00 ETH 2020-08-20 00:00:00 2021-12-01 00:00:00 ADA 2020-08-20 00:00:00 2021-12-01 00:00:00 DOT 2020-08-20 00:00:00 2021-12-01 00:00:00 SOL 2020-08-20 00:00:00 2021-12-01 00:00:00
gluonts.model.deepar._estimator.DeepAREstimator(alpha=0.0, batch_size=32, beta=0.0, cardinality=[2], cell_type="lstm", context_length=15, default_scale=None, distr_output=gluonts.mx.distribution.student_t.StudentTOutput(), dropout_rate=0.0, dropoutcell_type="ZoneoutCell", embedding_dimension=None, freq="1D", imputation_method=None, impute_missing_values=False, lags_seq=None, minimum_scale=1e-10, num_cells=8, num_imputation_samples=1, num_layers=2, num_parallel_samples=100, prediction_length=15, scaling=True, time_features=None, train_sampler=None, trainer=gluonts.mx.trainer._base.Trainer(add_default_callbacks=True, batch_size=None, callbacks=None, clip_gradient=10.0, ctx=None, epochs=200, hybridize=True, init="xavier", learning_rate=0.01, learning_rate_decay_factor=0.5, minimum_learning_rate=5e-05, num_batches_per_epoch=50, patience=5, weight_decay=1e-08), use_feat_dynamic_real=False, use_feat_static_cat=True, use_feat_static_real=False, validation_sampler=None)
100%|█| 50/50 [00:02<00:00, 24.12it/s, epoch=1/200, av 100%|█| 50/50 [00:02<00:00, 22.25it/s, epoch=2/200, av 100%|█| 50/50 [00:01<00:00, 26.13it/s, epoch=3/200, av 100%|█| 50/50 [00:01<00:00, 27.78it/s, epoch=4/200, av 100%|█| 50/50 [00:02<00:00, 24.19it/s, epoch=5/200, av 100%|█| 50/50 [00:01<00:00, 28.31it/s, epoch=6/200, av 60%|▌| 30/50 [00:01<00:00, 26.18it/s, epoch=7/200, av
--------------------------------------------------------------------------- KeyboardInterrupt Traceback (most recent call last) /tmp/ipykernel_2049/1003222560.py in <module> 16 ) 17 ---> 18 predictor = estimator.train(training_data=data) ~/stock-analysis/venv_39/lib/python3.9/site-packages/gluonts/mx/model/estimator.py in train(self, training_data, validation_data, num_workers, num_prefetch, shuffle_buffer_length, cache_data, **kwargs) 193 **kwargs, 194 ) -> Predictor: --> 195 return self.train_model( 196 training_data=training_data, 197 validation_data=validation_data, ~/stock-analysis/venv_39/lib/python3.9/site-packages/gluonts/mx/model/estimator.py in train_model(self, training_data, validation_data, num_workers, num_prefetch, shuffle_buffer_length, cache_data) 168 training_network = self.create_training_network() 169 --> 170 self.trainer( 171 net=training_network, 172 train_iter=training_data_loader, ~/stock-analysis/venv_39/lib/python3.9/site-packages/gluonts/mx/trainer/_base.py in __call__(self, net, train_iter, validation_iter) 434 logger.info(f"Epoch[{epoch_no}] Learning rate is {curr_lr}") 435 --> 436 epoch_loss = loop( 437 epoch_no, 438 train_iter, ~/stock-analysis/venv_39/lib/python3.9/site-packages/gluonts/mx/trainer/_base.py in loop(epoch_no, batch_iter, num_batches_to_use, is_training) 387 ) 388 --> 389 epoch_loss.update(None, preds=loss) 390 391 lv = loss_value(epoch_loss) ~/stock-analysis/venv_39/lib/python3.9/site-packages/mxnet/metric.py in update(self, _, preds) 1684 1685 for pred in preds: -> 1686 loss = ndarray.sum(pred).asscalar() 1687 self.sum_metric += loss 1688 self.global_sum_metric += loss ~/stock-analysis/venv_39/lib/python3.9/site-packages/mxnet/ndarray/ndarray.py in asscalar(self) 2583 raise ValueError("The current array is not a scalar") 2584 if self.ndim == 1: -> 2585 return self.asnumpy()[0] 2586 else: 2587 return self.asnumpy()[()] ~/stock-analysis/venv_39/lib/python3.9/site-packages/mxnet/ndarray/ndarray.py in asnumpy(self) 2561 """ 2562 data = np.empty(self.shape, dtype=self.dtype) -> 2563 check_call(_LIB.MXNDArraySyncCopyToCPU( 2564 self.handle, 2565 data.ctypes.data_as(ctypes.c_void_p), KeyboardInterrupt:
Running evaluation: 100%|██████████████████████████| 5/5 [00:00<00:00, 88.86it/s]
Till 2021-12-01 00:00:00: 0.12194796403249104 Till 2021-12-01 00:00:00: 0 0.094399 1 0.073585 2 0.136142 3 0.160227 4 0.145388 Name: MAPE, dtype: float64
Running evaluation: 100%|██████████████████████████| 5/5 [00:00<00:00, 94.63it/s]
Till 2021-11-16 00:00:00: 0.04887518763542176 Till 2021-11-16 00:00:00: 0 0.025752 1 0.043122 2 0.055691 3 0.078770 4 0.041042 Name: MAPE, dtype: float64
Running evaluation: 100%|██████████████████████████| 5/5 [00:00<00:00, 81.23it/s]
Till 2021-11-01 00:00:00: 0.06698968450228374 Till 2021-11-01 00:00:00: 0 0.021742 1 0.059310 2 0.042007 3 0.050233 4 0.161656 Name: MAPE, dtype: float64
--------------------------------------------------------------------------- NameError Traceback (most recent call last) /tmp/ipykernel_2049/3891290465.py in <module> 3 start_forecasting = df_dict['BTC'].index[-1] 4 ----> 5 index = pd.date_range(start_forecasting, periods=model_params['prediction_length'], freq='1D') 6 7 df_to_predict = pd.DataFrame(index=index) NameError: name 'model_params' is not defined
[NbConvertApp] Converting notebook crypto_analysis.ipynb to html [NbConvertApp] Writing 4238742 bytes to crypto_analysis.html